Predicting Maltreatment in Adolescents with Mentally Ill Parents: A Random Forest Tree Analysis

Area of Science:

• Psychiatry and Psychology
• Child Development
• Data Science in Healthcare

Background:

• Children of parents with mental illness are at elevated risk for developing mental health disorders and experiencing maltreatment.
• Early identification of maltreatment predictors in this vulnerable population is crucial for timely support and intervention.
• Existing research highlights the intergenerational impact of mental health conditions and adverse childhood experiences.

Purpose of the Study:

• To investigate the predictability of child maltreatment in families with a mentally ill parent.
• To evaluate the efficacy of machine learning models in predicting child maltreatment using parental and child self-reports.
• To identify key predictors for early risk assessment of maltreatment in at-risk children.

Main Methods:

• A random forest classifier was employed to analyze data from psychiatric inpatients and their children.
• Child maltreatment symptoms were assessed using the Childhood Trauma Questionnaire Short-Form.
• Three predictive models were developed: parent-estimated child trauma, parent-reported child maltreatment, and child self-reported maltreatment.

Main Results:

• Model 1 (parent-estimated trauma) achieved 76.62% accuracy with an Area Under the Curve (AUC) of .85.
• Model 3 (child self-reported maltreatment) demonstrated 73.68% accuracy with an AUC of .84.
• Model 2 (parental data for child self-assessed maltreatment) showed 68.42% accuracy and an AUC of .69, indicating moderate predictability.

Conclusions:

• Machine learning models offer moderate-to-good predictability for assessing child maltreatment in families with parental mental illness.
• Parental and child self-reports are valuable data sources for risk assessment, with parent-estimated trauma and child self-reports showing higher accuracy.
• These findings provide initial insights for developing targeted interventions and support systems for at-risk children in these families.